Mirror of the gdb mailing list
 help / color / mirror / Atom feed
From: Steffen Dettmer <steffen.dettmer@googlemail.com>
To: gdb@sourceware.org
Subject: Re: printing array in function
Date: Fri, 03 Sep 2010 10:24:00 -0000	[thread overview]
Message-ID: <AANLkTik9TFx-WaWgCg7=4DJn5wYAH5P0ojXwANLKUfh8@mail.gmail.com> (raw)
In-Reply-To: <4C80BA7A.3030309@lip6.fr>

On Fri, Sep 3, 2010, Nicolas Sabouret <Nicolas.Sabouret@lip6.fr> wrote:
> The problem we have is that arrays passed to functions are seen as
> pointers by gdb. Here is a simple example :
>
> 1:  void f(int tab[]) {
> 5:    int t[] = {-1,-1};
>
> (gdb) p tab     -> (int *) 0xbffff440

I think this is absolutely correct.

tab is an int pointer, the storage size of a referenced "array" isn't known.

> The only solution we found to display tab as an array is to use "p
> *tab@2", but this requires knowing the exact size of the array (2 in
> this example).

Yes, absolutely, same as in C:
------------------------------------------------------------------->8=======
   #include <stdio.h>

   void f(int tab[]) {
     tab[0] = 1;
     /* (%z is not known to all compilers) */
     printf("sizeof(tab) = %lu\n", (unsigned long)sizeof(tab));
   }
   int main() {
     int t[] = {-1,-1};
     printf("sizeof(t) = %lu\n", (unsigned long)sizeof(t));
     f(t);
     return 0;
   }
=======8<-------------------------------------------------------------------

produces

   sizeof(t) = 8
   sizeof(tab) = 4

so function f() also needs to know the size or number of
elements, even when a "size" is given:

   void f(int tab[2]) {
       assert(sizeof(tab) == sizeof(int*));
   }

to illustrate this, development rules may forbid array notation
for function notation instead of pointer, leading to:

   void f(int *tab) {
     tab[0] = 1;
   }

which, as far as I know, the same as above and should be portable.

> Our problem is that the gdb calls are integrated in a front-end for
> students (they do not type gdb commands directly) and that our frontend
> has no way of "guessing" what is the correct size for the array.

maybe by passing a size parameter (which is usually needed anyway
to be able to know the size of the array):

   void f(int *tab, size_t tab_size) {
     tab[0] = 1;
   }
   {
     int t[] = {-1,-1};
     f(t, sizeof(t)/sizeof(t[0]));
   }

and using tab_size in the frontend?

oki,

Steffen


  parent reply	other threads:[~2010-09-03 10:24 UTC|newest]

Thread overview: 7+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2010-09-03  9:06 Nicolas Sabouret
2010-09-03  9:16 ` Jan Kratochvil
2010-09-03  9:38   ` Nicolas Sabouret
2010-09-03 10:24 ` Steffen Dettmer [this message]
2010-09-03 16:21 ` Andreas Schwab
2010-09-03 19:18   ` Nicolas Sabouret
2010-09-03 23:10     ` Joachim Protze

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to='AANLkTik9TFx-WaWgCg7=4DJn5wYAH5P0ojXwANLKUfh8@mail.gmail.com' \
    --to=steffen.dettmer@googlemail.com \
    --cc=gdb@sourceware.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox